Provide ability to annotate subjects as the proband

[torstees]

Requester information

• Name: Eric Torstenson
• Affiliations: VUMC/Anvil

Request Details

• Purpose: For Family Relationship, the values are always in reference to the proband, but we aren't currently marking the proband in any way. Perhaps the absence of a family relationship is sufficient, but it seems better to be explicit
• Who it benefits: Anyone who has family relationships as part of their data
• Use case: For a dataset with extended family, there may be quite a few individuals which would make it a bit messier to identify the actual proband. Having a way to clearly identify which individual is a proband would could be very helpful.

[torstees]

An easy solution would be to hang a simple Observation off of the subject with a valueBoolean indicating True.

Some possible codes to consider for the observation:

• UMLS CUI: C1948021 (proband)
• NCI C64435 (proband)

HL7's does have ONSESELF as part of it's family member, PersonalRelationshipRoleType, value set

[RobertJCarroll]

That sounds reasonable to me. Regarding the concept, I'm not sure if it makes sense to use UMLS cuis, depends on what vocabularies we expect to be included. The UMLS concept is pretty much just the NCI term as well.